CareerPlanGet AI match score →

Software Engineer Distributed Systems Ai Infrastructure

💼 Full-time🗓 2026-07-27

Core

Building MLOps, edge-native infrastructure, and audit-grade data pipelines for high-stakes government and financial institutions.

Role type

Senior IC distributed systems engineer (MLOps & edge infrastructure)

Builds

Production MLOps pipelines, multi-tenant authorization systems, offline-first edge infrastructure, and immutable audit trails.

Domain

Public sector / Fintech / Insurance / Distributed Systems / MLOps

Deliverable

production ML models | infrastructure

Required skills

OS internals, distributed systems (consensus, CRDTs), LLM quantization, Go, Kubernetes (K3s), ClickHouse, Pinecone, Weaviate

Preferred skills

None stated

Technologies

Go, Kubernetes, K3s, ClickHouse, Pinecone, Weaviate, vLLM, TGI

Responsibilities

Own full model lifecycle (training, evaluation, versioning, deployment); design fine-grained authorization and multi-tenant isolation; architect edge-native systems with robust sync and conflict resolution; implement immutable logs and time-ordered event histories.

Seniority

Senior, hands-on IC

Rewrite
## About the Role Omara Technologies is building a data platform for highly regulated institutions in public service, insurance and private credit. We serve population-scale welfare systems, regulated fintech, and insurance operations - environments that demand correctness over convenience. Our platform is deployed within state government agencies to prevent fraud, optimize last-mile delivery, and automate clerical workloads at scale. ## The Team We're a small team from Yale, UPenn, and IIT. We value technical rigor, ownership, and the ability to ship production infrastructure under pressure. ## What You'll Build - MLOps for high-stakes environments. You'll own the full model lifecycle: training pipelines, evaluation frameworks, versioning, and deployment. - Permissions and access control. You'll design fine-grained authorization and multi-tenant isolation for government-scale deployments. - Edge-native infrastructure. You'll architect systems that survive unreliable networks using robust sync, conflict resolution, and offline-first workflows (local LLMs, vLLM/TGI optimization). - Audit-grade data pipelines. You'll implement strict chain-of-custody: immutable logs, evidence-grade trails, and time-ordered event histories that hold up under sovereign-scale scrutiny. ## Requirements - Deep understanding of OS internals, networking, and distributed systems (consensus protocols, CRDTs, or event-sourcing). - Hands-on experience with LLM quantization, evaluation frameworks, and production-grade inference optimization. - Proficiency in Go, Kubernetes (K3s for edge), and specialized data stores (ClickHouse, Pinecone/Weaviate). - Degree in Computer Science, Mathematics, Physics, or a related quantitative field from a top-tier institution (IIT, IIIT-H or equivalent). ## Who You Are - An owner: You move from ambiguity to specification without a roadmap. You own the uptime, security, and reliability of what you build. - A polyglot: You're comfortable across the stack - ML-adjacent work, backend infrastructure, high-fidelity UI, mobile interfaces. - Resilient: You prefer the simplest system that works, then make it bulletproof. You thrive in high-intensity, early-stage environments. ## How to Apply Email [email protected] or apply via Wellfound. Include your GitHub or technical portfolio and a brief note on the most complex system you've owned in production.
Sourced via wellfound · Listed on CareerPlan, which tracks 70,000+ jobs from 20+ sources.
Apply on Wellfound ↗